ABSTRACT The Specific Aim of this proposal is to test the feasibility of developing a web-based peer mentor training and mentor-mentee matching system to help patients better manage their diabetes through mentoring. Effective self-management of diabetes is the key to improving diabetes outcomes and reducing costs. Self-management includes medication adherence, self-monitoring, healthy eating, and physical activity. Yet, many patients do not effectively sustain these required behaviors. Trained peer mentors successfully managing their own diabetes are an effective resource to help struggling patients with uncontrolled diabetes. Randomized clinical trials led by our team (Thom et al. 2013 and Long et al. 2012) found that peer-to-peer mentoring can result in a clinically significant reduction of HbA1c levels (one point), thus improving the quality of life for patients. However, two major challenges prevent widespread adoption of this cost-effective, evidence-based intervention: (1) training peer mentors currently requires live training, which is expensive and not scalable and (2) current methods for appropriate matching of peers and peer mentors are inefficient and time-consuming, and thus not scalable. To overcome these weaknesses, we are developing a web-based program for online training of peer mentors and for assisted-matching of mentors to mentees. The training program and matching criteria will be based on methods previously validated through two randomized clinical trials led by our team. This program is expected to allow for clinical translation of our initial research findings. To achieve our Phase I Specific Aim, we will carry out two Tasks: Task 1: Develop and test the feasibility of remote peer mentor training. Test of Feasibility Task 1: At least 70% of mentors quantitatively show preparedness in a written and phone- based, scenario-driven practicum with our research team (90%+ correct), at least 70% of mentors are satisfied with the training (score at 3 or 4 on a 0-4 Likert scale with 4 being completely satisfied), and there exists a significant improvement in both written and oral exams for mentors from baseline to training completion. Task 2: Develop and evaluate usability of a web-based system to facilitate mentor-mentee matching by medical staff. Test of Feasibility Task 2: At least 70% of administrators and patients are satisfied with the system (Score at 3 or 4 on a 0-4 Likert scale with 4 being completely satisfied). Phase II Plans: Fully develop training modules for peer mentors and make additional improvements, as informed by Phase I; conduct a multi-arm randomized clinical trial to evaluate the efficacy of mentoring on quality of life and diabetes management when those mentors have been trained and matched to mentees using our scalable cloud based training and matching system.